EUVIMEDAtklāt EUVIMED
Systematische ÜbersichtMetaanalyseÜbersichtsarbeitOpen access

Machine Learning Models for Glycemic Prediction in Type 1 and Type 2 Diabetes: Systematic Review and Meta-Analysis

Machine-Learning-Modelle zur glykämischen Vorhersage bei Typ-1- und Typ-2-Diabetes: systematische Übersichtsarbeit und Metaanalyse

Haoran Yang, Diliyaer Abudukeremu, Yuhan Shi, Li Fang, Lei Li, Wenzheng Gu, Yunqi Jia, Rongfang He

Journal of Medical Internet Research · 2026-11-09 · 28 · e100653-e100653

DOI: 10.2196/100653 ↗

Auf einen Blick

Die Übersichtsarbeit fand Hinweise auf Vorhersageleistung von Machine-Learning-Modellen für glykämische Ereignisse und Glukosewerte. Die Evidenz war jedoch heterogen und nur ein kleiner Teil quantitativ auswertbar.

Fragestellung

Wie gut sagen Machine-Learning-Modelle glykämische Ereignisse oder Glukosewerte bei Menschen mit Typ-1-, Typ-2- oder gemischtem Diabetes voraus?

Methode

Systematische Übersichtsarbeit und Metaanalyse; Suche in 9 englisch- und chinesischsprachigen Datenbanken bis 12. April 2026; getrennte Analyse von kontinuierlicher Vorhersage und binärer Ereignisvorhersage; PROBAST-Bewertung.

Ergebnisse

299 Berichte mit 270 Studien; 19 Berichte für die quantitative Synthese. Gepooltes RMSE 1,18 mmol/L (95 %-KI 1,01–1,34; 37 Effekte), MAE 0,90 mmol/L (95 %-KI 0,78–1,02; 20 Effekte), AUC 0,926 (95 %-KI 0,877–0,957).

Einschränkungen

Heterogenität, hohes Verzerrungsrisiko, unvollständige Unsicherheitsangaben sowie geringe externe und prospektive Validierung.

Schlussfolgerung der Autor:innen

Die Studien zeigen Vorhersage- und Diskriminationssignale, aber die Heterogenität, das hohe Bias-Risiko und die begrenzte Validierung schließen belastbare Benchmarks aus.

Kernaussagen

  • RMSE gepoolt: 1,18 mmol/L
  • AUC gepoolt: 0,926
  • Nur 4,0 % externe und 0,7 % prospektive Validierung

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

At a glance

Systematische Übersicht in Journal of Medical Internet Research, published online on 2026-11-09. Written by Haoran Yang and 7 co-authors. Institution: Department of Pathology, The First Affiliated Hospital, Hengyang Medical School, University of South China, No. 69, Chuanshan Road, Hengyang, 421001,…. Major MEDLINE topics: Machine Learning, Diabetes Mellitus, Type 1, Diabetes Mellitus, Type 2, Blood Glucose. 61 references. The full text is freely available in PubMed Central.

Auf einen BlickAbstraktsSubjectsBibliogrāfiskie datiCiteRelated studies

Abstract (original language)

Background

Machine learning models are increasingly used for glycemic event prediction and continuous glucose forecasting, but targets, populations, horizons, validation, and uncertainty reporting remain heterogeneous.

Objective

This study aimed to evaluate machine learning models for glycemic prediction in type 1, type 2, and mixed diabetes populations while distinguishing the full evidence map from the quantitatively analyzable subset.

Methods

We searched 9 English- and Chinese-language databases from inception to April 12, 2026, for studies predicting future glycemic events or glucose values in people with diabetes. Continuous forecasting and binary event prediction were analyzed separately. Root-mean-square error (RMSE) was the principal continuous end point; mean absolute error was supportive. Area under the receiver operating characteristic curve was synthesized using a Bayesian random-effects model because analyzable 2×2 data were sparse. Two reviewers (HY and DA) independently applied PROBAST (Prediction Model Risk of Bias Assessment Tool) to 19 quantitative-source reports (28 assessment units).

Results

We identified 787 records, assessed 310 full texts, and included 299 reports representing 270 studies; 19 reports contributed to quantitative synthesis. The review included 128 type 1, 26 type 2, 78 mixed, and 38 unclear-diabetes-type studies. Pooled RMSE was 1.18 mmol/L (95% CI 1.01‐1.34; 37 effects), supportive mean absolute error was 0.90 mmol/L (95% CI 0.78‐1.02; 20 effects), and Bayesian pooled area under the receiver operating characteristic curve was 0.926 (95% credible interval 0.877‐0.957). Supportive sensitivity and specificity were 0.77 and 0.80. Of 28 PROBAST units, 24 were high risk, 2 unclear, and 2 low. Post hoc exploratory within-report analysis estimated an RMSE increase of 0.481 mmol/L per additional 30 minutes (95% CI 0.213‐0.749), but overlap sensitivity reduced certainty. In the 270-study clinical readiness audit, external validation occurred in 4.0% (10/251), prospective evaluation in 0.7% (2/270), calibration in 4.8% (13/270), decision-curve or net-benefit analysis in 2.6% (7/270), and implementation or human-factors evaluation in 2.6% (7/270).

Conclusions

Glycemic prediction studies show forecasting and discrimination signals, but heterogeneity, high risk of bias, incomplete uncertainty reporting, and limited external and prospective validation preclude deployable benchmarks. Pooled estimates should be viewed as summaries of analyzable literature. Future studies need transparent horizon, threshold, calibration, validation, uncertainty, and implementation reporting.

Subjects

MeSH terms (MEDLINE)

Machine LearningDiabetes Mellitus, Type 1Diabetes Mellitus, Type 2Blood GlucoseHumansBayes TheoremPredictive Learning ModelsPrediction Algorithms

Author keywords

artificial intelligenceblood glucose forecastingcontinuous glucose monitoringdiabetes mellitushyperglycemiahypoglycemiamachine learningmeta-analysissystematic reviewtype 1 diabetestype 2 diabetes

Substances

Blood Glucose

Authors and institutions

  • Department of Pathology, The First Affiliated Hospital, Hengyang Medical School, University of South China, No. 69, Chuanshan Road, Hengyang, 421001, China, +86 18773464745.
  • Tianjin Medical University Cancer Institute and Hospital, Tianjin, China.
  • The Second Clinical College of Tianjin Medical University Second Hospital, Tianjin, China.
  • Meiwan Community Health Service Station (Municipal Party Committee Clinic), Yongzhou, China.
  • Cai Township Health Center, Yongzhou, China.
  • First Teaching Hospital of Tianjin University of Traditional Chinese Medicine, National Clinical Research Center for Chinese Medicine, Tianjin, China.

Cite

APA

Yang, H., Abudukeremu, D., Shi, Y., Fang, L., Li, L., Gu, W., Jia, Y., & He, R. (2026). Machine Learning Models for Glycemic Prediction in Type 1 and Type 2 Diabetes: Systematic Review and Meta-Analysis. Journal of Medical Internet Research, 28, e100653-e100653. https://doi.org/10.2196/100653

Vancouver

Yang H, Abudukeremu D, Shi Y, Fang L, Li L, Gu W, et al. Machine Learning Models for Glycemic Prediction in Type 1 and Type 2 Diabetes: Systematic Review and Meta-Analysis. Journal of Medical Internet Research. 2026;28:e100653-e100653. doi:10.2196/100653

BibTeX
@article{Yang2026,
  author = {Yang, Haoran and Abudukeremu, Diliyaer and Shi, Yuhan and Fang, Li and Li, Lei and Gu, Wenzheng and Jia, Yunqi and He, Rongfang},
  title = {Machine Learning Models for Glycemic Prediction in Type 1 and Type 2 Diabetes: Systematic Review and Meta-Analysis},
  journal = {Journal of Medical Internet Research},
  year = {2026},
  volume = {28},
  pages = {e100653--e100653},
  doi = {10.2196/100653}
}
Start for free →